Appendix C: Content Analysis
After an informal examination of the material, definitions
of the score categories were agreed upon and noted in
writing. Several magazines were scored jointly, and
discussions of borderline cases resulted in slight adjustments
of our definitions. Scoring was done on prepared score
sheets, with one sheet for each magazine series. All
photographs in the magazine were scored. When either one
of us encountered a photograph which might have been
interpreted in more than one way, we consulted each
other. The scoring was done over a period of several
consecutive days to ensure consistency and because we wanted
to return the material to its owners as soon as possible.
The total number of photographs analyzed was 10,370.
Of these, 4,630 were found in girl magazines and 5,740 in boy
magazines. The following variables were registered:
(1) |
gender, |
(2) |
country of origin, |
(3) |
number of people in the photograph, |
(4) |
the hardness index. |
The following aspects were analyzed:
(1) |
the frequency distribution, with X2
test. |
(2) |
the hardness of the pornography, calculated as the
weighted sum of the photographs, divided by the
number of photographs. The weighting was given
by the hardness index as follows: clothed = 0, naked
= 1, posed = 2, contact = 3, oral= 4, genital = 5,
bizarre = 6. We used the non-parametric
Wilcoxon rank sum test for two independent variables
and the Wilcoxon signed rank test for two
independent variables. [*199] |
The chosen weights cannot be regarded as having any
absolute value, since the choice of intervals is
arbitrary. It is obvious that the category
"bizarre" should carry a heavier weight than
"genital," but how much higher is not possible to
determine by objective means.
Boys versus Girls
The frequency distribution of hardness for boy and girl
magazines is shown in Figure C1.
The graphic is described in the body of the text and shown
in Table 2 on page 70. The distribution for sex is
significant: X2 (df = 6) is 1351 (p <
0.0001).
Table 2- Division of Photographs
in the Sample
|
Boys
|
Girls
|
Total
|
0. Clothed |
800(17.3%) |
85 (1.5%) |
885 (8.5%) |
1. Naked |
1120(24.2%) |
917 (16.2%) |
2037(19.6%) |
2. Posed |
1249(27.2%) |
2155 (37.5%) |
3404(32.8%) |
3. Contact |
1030(22.2%) |
1096(19.1%) |
2126 (20.5%) |
4. Oral |
218 (4.7%) |
490(8.5%) |
708 (6.8%) |
5. Genital |
192(4.1%) |
922(16.1%) |
1114(10.7%) |
6. Bizarre |
21(0.5%) |
75 (1.3%) |
96(0.9%) |
|
|
|
|
Total |
4630(100%) |
5740(100%) |
10370(100%) |
|
The average hardness index for girl and boy magazines are
respectively 2.71 and 1.87. The average for all the
magazines was 2.33. The difference in hardness index
between the two groups, tested with Wilcoxon rank sum test is
significant: X2 (girls) = 214, X2
(boys) = 75; p < 0.01.
The Country of Origin
The photographs were divided as seen in Table C1. [*200]
Table C1
Hardness Distribution for Country of Origin
|
Denmark
|
Netherlands
|
Germany
|
Sweden
|
0. clothed
|
141 |
62 |
384 |
25 |
1. naked
|
434 |
618 |
555 |
56 |
2. posed
|
1552 |
1062 |
300 |
88 |
3. contact
|
1202 |
600 |
54 |
75 |
4. oral
|
335 |
270 |
0 |
13 |
5. genital
|
477 |
496 |
0 |
26 |
6. bizarre
|
46 |
22 |
0 |
0 |
|
|
|
|
|
Total:
|
4187 |
3130 |
1293 |
283 |
Average:
|
2.66 |
2.63 |
1.02 |
2.26 |
|
The distribution of hardness for the different countries of
origin was significantly different: X2 (df =
18) is 2409 (p < 0.0001). "Average" in the
table above refers to the average hardness index for the
photographs of the respective countries. Testing for
significance between individual countries was not done, since
the sample is not large enough to yield meaningful results.
Persons in the Photographs
Finally, a distinction was made between photographs
depicting one child, photographs depicting two or more
children, and photographs in which children with adults
appeared. Significance testing was carried out for all
the magazines, the girl magazines and boy magazines. [*201]
In a certain sense, an adult is present in all the
photographs: the photographer. For naturist photographs,
in which the child is not conscious of the picture being
taken, this obviously plays no role in determining the
demeanor of the child. For many of the photographs, the
child is posing, facing the camera, and clearly conscious of
the pictures being taken. The photographer, presumably
an adult, is clearly present by implication. Because of
obvious difficulties we have restricted this analysis to the
objective picture content. Adults who do not appear in
the photograph are not counted as present.
We considered making a distinction between
"pornographic" and "sexual" activity,
defined as follows: Pornographic activity is directed at
arousing the viewer of the photograph, so that display of the
genitals and posing provocatively would be defined as
pornographic. On the other hand, sexual activity is
focused on the arousal of the child. Masturbation
scenes, in which the child is absorbed in the sexual activity
without apparent attention to the photographer, would be
defined as sexual. As a general trend the girl
photographs appeared more often pornographic and the boy
photographs more often sexual. We did not attempt to
quantify this because of the difficulty finding an objective
criterion for determining the thoughts of the child at the
time the picture was taken.
Girl Magazines
The number of photographs are given in Table C2. The
letters represent one child (A), two or more children (B) and
one or more children with an adult (C). On more than
half of the photographs in these magazines only one girl can
be seen alone. The number on which more than one girl is
seen, without adults, is the lowest.
Table C2
Girl Magazines — Number of Photographs with
Different Hardness Classifications
|
A
|
B
|
C
|
0. clothed
|
78 |
4 |
3 |
1. naked
|
675 |
184 |
58 |
2. posed
|
1781 |
218 |
156 |
3. contact
|
440 |
200 |
456 |
4. oral
|
0 |
55 |
435 |
5. genital
|
234 |
80 |
608 |
6. bizarre
|
7 |
2 |
66 |
|
|
|
|
Total:
|
3215 |
743 |
1782 |
Average:
|
2.11 |
2.49 |
3.88 |
|
The distribution of the photographs over the categories of
hardness is significantly different for the three groups: X2
(df = 12) is 2548 (p < 0.0001).
The majority of the photographs without adults present were
mild, being mostly natural naked and posed naked photographs,
while the majority of photographs which include adults have a
heavy pornographic character. Category five shows a jump
representing for the large part display of the genitals, such
as parting of the labia with the hands, though some of them
showed masturbation, not only showing the child explicitly
engaged in pornographic activity but also conscious of the
photographer.
On the basis of these indices the difference in hardness
for the groups A, B and C were tested with the Wilcoxon signed
rank test. The difference in hardness between A and B
photographs is significant at p = 0.0266. The difference
between the B and C photographs is significant at p <
0.0001.
Boy Magazines
The division of the boy photographs over the three groups
is shown in Table C3.
Table C3
Boy Magazines — Number of Photographs with
Different Hardness Classifications
|
A
|
B
|
C
|
0. clothed
|
647 |
112 |
41 |
1. naked
|
805 |
295 |
20 |
2. posed
|
1046 |
148 |
55 |
3. contact
|
161 |
674 |
195 |
4. oral
|
0 |
145 |
73 |
5. genital
|
0 |
52 |
140 |
6. bizarre
|
0 |
0 |
21 |
|
|
|
|
Total:
|
2659 |
1426 |
545 |
Average:
|
1.27 |
2.42 |
3.36 |
|
The X2 test for independence is
significant: X2 (df =12) is 26l7 (p <
0.0001).
In the boy photographs, as in the girl photographs, more
than half of the photographs consisted of one boy alone,
frequently posing naked, but mostly natural photographs and a
remarkably large proportion of portrait and fully clothed
photographs. The number of photographs in which an adult
appears is also much lower than in the girl magazines.
The boy photographs include many more non-pornographic and
mild photographs, but, at the same time, a greater proportion
of the B category fall in the middle range, and a greater
proportion of photographs with an adult fell in the three
hardest categories, which is reflected in the average values.
One can speculate over the possible reasons for these
differences. The increased proportion of boys found in
the B group, and the higher average hardness index for this
group, may be the result of the greater number of
possibilities, both photographic and action, when two boys get
together compared to the possibilities that exist when two
girls get together. Boys, after all, have both the
equipment and inclination to act out their sexuality in more
obvious ways. Of the boys photographed with adults the
proportion engaged in the hardest activities with adults was
slightly lower than for girls, nevertheless, some of the boy
pictures, which we placed in the bizarre category, showed
violent rape, including some in which the boy was screaming in
pain. These differences mean that comparison of the
groups between sexes should be done with caution, for the
meaning of group B and group C, as defined by us, may not be
exactly the same for boys and for girls.
On the basis of the hardness index of the boy photographs,
the difference in the distribution of hardness was tested with
the Wilcoxon signed rank test.
The difference between A and B photographs is significant
(p < 0.0001). The difference between B and C is also
significant (p = 0.0078).
All Magazines
The division of all 10,370 photographs over three groups is
seen in Table C4. More than half of the photographs show
only one child, and the remainder is divided equally between
photographs with two or more children and photographs with
adults present. The majority (61%) fall into the three
lightest categories, and in the photographs without adults
present it is 75%, while with adults it is only 14%.
Table C4
All Magazines — Number of Photographs with
Different Classifications
|
A
|
B
|
C
|
0. clothed
|
725 |
116 |
44 |
1. naked
|
1480 |
479 |
78 |
2. posed
|
2827 |
366 |
211 |
3. contact
|
601 |
874 |
651 |
4. oral
|
0 |
200 |
508 |
5. genital
|
234 |
132 |
748 |
6. bizarre
|
7 |
2 |
87 |
|
|
|
|
Total:
|
5874 |
2169 |
2327 |
Average:
|
1.73 |
2.45 |
3.76 |
|
The distribution within the three categories was
significantly different for all three groups. X2
(df = 12) is 5155 (p < 0.0001). The difference
between group A and group B, tested with the Wilcoxon signed
rank test was significant (p < 0.0001) and the same for the
B and C groups (p < 0.0001).
It would be a mistake to extrapolate from the photographs
to the kinds of activities that children engage in with each
other and with adults when sexual contacts take place.
Persons engaged in violent abuse may be less likely to record
the events for fear of exposure. Likewise, relationships
in which love and mutual respect characterize the encounters
may also be less likely to appear in published pornography,
since knowing publication of such pictures is hard to
reconcile with respect for the child. The differences
and parallels between that which is seen on the photographs
and that which typically occurs among children and between
adults and children when sexual relationships take place
cannot be determined from the photographs alone.